scholarly journals Segmentation of Overlapping Grape Clusters Based on the Depth Region Growing Method

Electronics ◽  
2021 ◽  
Vol 10 (22) ◽  
pp. 2813
Author(s):  
Yun Peng ◽  
Shengyi Zhao ◽  
Jizhan Liu
Keyword(s):  
Region Of Interest ◽  
Region Growing ◽  
Difficult Problem ◽  
Depth Image ◽  
Practical Application ◽  
Simple Method ◽  
Learning Network ◽  
Deep Learning Network ◽  
Primary Problem ◽  
Rgb Image

Accurately extracting the grape cluster at the front of overlapping grape clusters is the primary problem of the grape-harvesting robot. To solve the difficult problem of identifying and segmenting the overlapping grape clusters in the cultivation environment of a trellis, a simple method based on the deep learning network and the idea of region growing is proposed. Firstly, the region of grape in an RGB image was obtained by the finely trained DeepLabV3+ model. The idea of transfer learning was adopted when training the network with a limited number of training sets. Then, the corresponding region of the grape in the depth image captured by RealSense D435 was processed by the proposed depth region growing algorithm (DRG) to extract the front cluster. The depth region growing method uses the depth value instead of gray value to achieve clustering. Finally, it fils the holes in the clustered region of interest, extracts the contours, and maps the obtained contours to the RGB image. The images captured by RealSense D435 in a natural trellis environment were adopted to evaluate the performance of the proposed method. The experimental results showed that the recall and precision of the proposed method were 89.2% and 87.5%, respectively. The demonstrated performance indicated that the proposed method could satisfy the requirements of practical application for robotic grape harvesting.

scholarly journals A New Deep Learning Network for Automatic Bridge Detection from SAR Images Based on Balanced and Attention Mechanism

2020 ◽  
Vol 12 (3) ◽  
pp. 441
Author(s):  
Lifu Chen ◽  
Ting Weng ◽  
Jin Xing ◽  
Zhouhao Pan ◽  
Zhihui Yuan ◽  
...  
Keyword(s):  
Deep Learning ◽  
Region Of Interest ◽  
Recall Rate ◽  
Attention Mechanism ◽  
Sar Images ◽  
Learning Network ◽  
Feature Pyramid ◽  
Pooling Strategy ◽  
Classification And Regression ◽  
Deep Learning Network

Bridge detection from Synthetic Aperture Radar (SAR) images has very important strategic significance and practical value, but there are still many challenges in end-to-end bridge detection. In this paper, a new deep learning-based network is proposed to identify bridges from SAR images, namely, multi-resolution attention and balance network (MABN). It mainly includes three parts, the attention and balanced feature pyramid (ABFP) network, the region proposal network (RPN), and the classification and regression. First, the ABFP network extracts various features from SAR images, which integrates the ResNeXt backbone network, balanced feature pyramid, and the attention mechanism. Second, extracted features are used by RPN to generate candidate boxes of different resolutions and fused. Furthermore, the candidate boxes are combined with the features extracted by the ABFP network through the region of interest (ROI) pooling strategy. Finally, the detection results of the bridges are produced by the classification and regression module. In addition, intersection over union (IOU) balanced sampling and balanced L1 loss functions are introduced for optimal training of the classification and regression network. In the experiment, TerraSAR data with 3-m resolution and Gaofen-3 data with 1-m resolution are used, and the results are compared with faster R-CNN and SSD. The proposed network has achieved the highest detection precision (P) and average precision (AP) among the three networks, as 0.877 and 0.896, respectively, with the recall rate (RR) as 0.917. Compared with the other two networks, the false alarm targets and missed targets of the proposed network in this paper are greatly reduced, so the precision is greatly improved.

Real-time bi-directional people counting using an RGB-D camera

Sensor Review ◽  
2021 ◽  
Vol 41 (4) ◽  
pp. 341-349
Author(s):  
Wahyu Rahmaniar ◽  
W.J. Wang ◽  
Chi-Wei Ethan Chiu ◽  
Noorkholis Luthfil Luthfil Hakim
Keyword(s):  
Real Time ◽  
Region Of Interest ◽  
Computation Time ◽  
Depth Image ◽  
Fast Computation ◽  
Content Type ◽  
People Counting ◽  
Rgb Images ◽  
Precision Rate ◽  
Rgb Image

Purpose The purpose of this paper is to propose a new framework and improve a bi-directional people counting technique using an RGB-D camera to obtain accurate results with fast computation time. Therefore, it can be used in real-time applications. Design/methodology/approach First, image calibration is proposed to obtain the ratio and shift values between the depth and the RGB image. In the depth image, a person is detected as foreground by removing the background. Then, the region of interest (ROI) of the detected people is registered based on their location and mapped to an RGB image. Registered people are tracked in RGB images based on the channel and spatial reliability. Finally, people were counted when they crossed the line of interest (LOI) and their displacement distance was more than 2 m. Findings It was found that the proposed people counting method achieves high accuracy with fast computation time to be used in PCs and embedded systems. The precision rate is 99% with a computation time of 35 frames per second (fps) using a PC and 18 fps using the NVIDIA Jetson TX2. Practical implications The precision rate is 99% with a computation time of 35 frames per second (fps) using a PC and 18 fps using the NVIDIA Jetson TX2. Originality/value The proposed method can count the number of people entering and exiting a room at the same time. If the previous systems were limited to only one to two people in a frame, this system can count many people in a frame. In addition, this system can handle some problems in people counting, such as people who are blocked by others, people moving in another direction suddenly, and people who are standing still.

scholarly journals A Simple Method for Automatic 3D Reconstruction of Coronary Arteries From X-Ray Angiography

2021 ◽  
Vol 12 ◽  
Author(s):  
Minki Hwang ◽  
Sa-Bin Hwang ◽  
Hyosang Yu ◽  
Jaehyeok Kim ◽  
Daehyun Kim ◽  
...  
Keyword(s):  
Deep Learning ◽  
Coronary Arteries ◽  
Distal Region ◽  
Proximal Region ◽  
Automatic Identification ◽  
Success Rates ◽  
Simple Method ◽  
X Ray ◽  
Learning Network ◽  
Deep Learning Network

Automatic three-dimensional (3-D) reconstruction of the coronary arteries (CA) from medical imaging modalities is still a challenging task. In this study, we present a deep learning-based method of automatic identification of the two ends of the vessel from X-ray coronary angiography (XCA). We also present a method of using template models of CA in matching the two-dimensional segmented vessels from two different angles of XCA. For the deep learning network, we used a U-net consisting of an encoder (Resnet) and a decoder. The two ends of the vessel were manually labeled to generate training images. The network was trained with 2,342, 1,907, and 1,523 labeled images for the left anterior descending (LAD), left circumflex (LCX), and right coronary artery (RCA), respectively. For template models of CA, ten reconstructed 3-D models were averaged for each artery. The accuracy of correspondence using template models was compared with that of manual matching. The deep learning network pointed the proximal region (20% of the total length) in 97.7, 97.5, and 96.4% of 315, 201, and 167 test images for LAD, LCX, and RCA, respectively. The success rates in pointing the distal region were 94.9, 89.8, and 94.6%, respectively. The average distances between the projected points from the reconstructed 3-D model to the detector and the points on the segmented vessels were not statistically different between the template and manual matchings. The computed FFR was not significantly different between the two matchings either. Deep learning methodology is feasible in identifying the two ends of the vessel in XCA, and the accuracy of using template models is comparable to that of manual correspondence in matching the segmented vessels from two angles.

scholarly journals UAVs in rail damage image diagnostics supported by deep-learning networks

2021 ◽  
Vol 11 (1) ◽  
pp. 339-348
Author(s):  
Piotr Bojarczak ◽  
Piotr Lesiak
Keyword(s):  
Deep Learning ◽  
Head Width ◽  
Learning Networks ◽  
Image Brightness ◽  
Learning Network ◽  
Rail Head ◽  
Image Recording ◽  
Python Language ◽  
Deep Learning Network ◽  
Efficiency Rate

Abstract The article uses images from Unmanned Aerial Vehicles (UAVs) for rail diagnostics. The main advantage of such a solution compared to traditional surveys performed with measuring vehicles is the elimination of decreased train traffic. The authors, in the study, limited themselves to the diagnosis of hazardous split defects in rails. An algorithm has been proposed to detect them with an efficiency rate of about 81% for defects not less than 6.9% of the rail head width. It uses the FCN-8 deep-learning network, implemented in the Tensorflow environment, to extract the rail head by image segmentation. Using this type of network for segmentation increases the resistance of the algorithm to changes in the recorded rail image brightness. This is of fundamental importance in the case of variable conditions for image recording by UAVs. The detection of these defects in the rail head is performed using an algorithm in the Python language and the OpenCV library. To locate the defect, it uses the contour of a separate rail head together with a rectangle circumscribed around it. The use of UAVs together with artificial intelligence to detect split defects is an important element of novelty presented in this work.

scholarly journals 2D–3D reconstruction of distal forearm bone from actual X-ray images of the wrist using convolutional neural networks

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Ryoya Shiode ◽  
Mototaka Kabashima ◽  
Yuta Hiasa ◽  
Kunihiro Oka ◽  
Tsuyoshi Murase ◽  
...  
Keyword(s):  
Wrist Joint ◽  
High Accuracy ◽  
Training Data ◽  
Small Data ◽  
Data Set ◽  
Accuracy Estimation ◽  
X Ray ◽  
Learning Network ◽  
Forearm Bone ◽  
Deep Learning Network

AbstractThe purpose of the study was to develop a deep learning network for estimating and constructing highly accurate 3D bone models directly from actual X-ray images and to verify its accuracy. The data used were 173 computed tomography (CT) images and 105 actual X-ray images of a healthy wrist joint. To compensate for the small size of the dataset, digitally reconstructed radiography (DRR) images generated from CT were used as training data instead of actual X-ray images. The DRR-like images were generated from actual X-ray images in the test and adapted to the network, and high-accuracy estimation of a 3D bone model from a small data set was possible. The 3D shape of the radius and ulna were estimated from actual X-ray images with accuracies of 1.05 ± 0.36 and 1.45 ± 0.41 mm, respectively.

scholarly journals BioECG: Improving ECG Biometrics with Deep Learning and Enhanced Datasets

2021 ◽  
Vol 11 (13) ◽  
pp. 5880
Author(s):  
Paloma Tirado-Martin ◽  
Raul Sanchez-Reillo
Keyword(s):  
Deep Learning ◽  
Temporal Variations ◽  
Physical Activities ◽  
Learning Tools ◽  
Biometric Recognition ◽  
Learning Network ◽  
Deep Learning Network ◽  
Electrocardiogram Ecg

Nowadays, Deep Learning tools have been widely applied in biometrics. Electrocardiogram (ECG) biometrics is not the exception. However, the algorithm performances rely heavily on a representative dataset for training. ECGs suffer constant temporal variations, and it is even more relevant to collect databases that can represent these conditions. Nonetheless, the restriction in database publications obstructs further research on this topic. This work was developed with the help of a database that represents potential scenarios in biometric recognition as data was acquired in different days, physical activities and positions. The classification was implemented with a Deep Learning network, BioECG, avoiding complex and time-consuming signal transformations. An exhaustive tuning was completed including variations in enrollment length, improving ECG verification for more complex and realistic biometric conditions. Finally, this work studied one-day and two-days enrollments and their effects. Two-days enrollments resulted in huge general improvements even when verification was accomplished with more unstable signals. EER was improved in 63% when including a change of position, up to almost 99% when visits were in a different day and up to 91% if the user experienced a heartbeat increase after exercise.

Sign in / Sign up

Export Citation Format

Share Document